No description available.

No description available.

The need for automated fault detection methods has increased in line with the complexity of processing plant technology and their control systems. Fast and accurate fault detection and isolation (FDI) is essential if a controller is to be effective in a supervisory role. This thesis is concerned with developing an FDI system based upon artificial neural network techniques. The artificial neural network (ANN) is a mechanism based upon the concepts of information processing within the brain, and consequently has the ability to self adjust, or learn about a given problem domain. It can thus be utilised in currently favoured model-based FDI systems with the advantage that it can learn process dynamics by being presented examples of process input-output pairs without the need for traditional mathematically complex models. Similarly, ANNs can be taught to classify characteristics in the residual (or plant-model difference) signal without the necessity of constructing the types of filter used in more classical solutions. Initially, a class of feedforward neural network called the multilayer perceptron (MLP) is used to model mathematically simulated linear and nonlinear plants in order to demonstrate their abilities in this field, as well as investigating the consequence of parameter variation on model effectiveness and how the model can be utilised in a model-based FDI system. A principle aim of this research is to demonstrate the ability of the system to work online and in real-time on genuine industrial processes, and the plant nominated as a test bed - the Unilever Automated Freezer (UAF) - is introduced. The UAF, being a time-varying system, requires a novel system identification approach which has resulted in a number of cascaded MLPs to model the various stages in the phased startup of the process. In order to reduce model mismatch to a minimum, it was necessary to develop an effective switching mechanism between one MLP in the cascade and the next. Attempts using a rule-based switching mechanism, a simple MLP switch and an error based switching mechanism were made, before a solution incorporating a genetic algorithm and an MLP network was developed which had the capability of learning the optimum switching points. After the successful development of the model, a series of MLPs were trained to recognise the characteristics of a number of faults within the residual signals. Problems involving false alarms between certain faults were reduced by the introduction of templates - or information pertaining to when a particular fault was most evident in the residuals. The final solution consisting of an MLP Cascade model and fault isolation MLPs is essentially generic for this class of time-varying system, and the results achieved on the UAF were far superior to those of the currently used FDI system without the need for any extra sensory information. The MLP Cascade and associated switching device together with the development of an online real-time FDI system for a time-varying piece of industrial machinery, are deemed to be original contributions to knowledge.

The continuing failure of many CAPM implementations in the UK, despite the availability of a large and growing supply of software, gave rise to concern within the academic and user community and highlighted a need for a concerted research effort into the causes of failure. The Science and Engineering Research Council responded to this need by sponsoring a major research initiative into CAPM through its ACME (Application of Computers to Manufacturing Engineering) Directorate. The findings reported in this thesis result from the work carried out for ACME by the author as part of the joint Polytechnic South West/Sheffield Business School research programme under the direction of professors D R Hughes, I S Smith and D R Tranfield. The extent of the work included surveying a large number of manufacturing firms, interviewing suppliers of CAPM systems and services and analysing the data collected from these activities. This resulted in the identification of a number of causes of CAPM failure. A major component of the work then concerned developing an appropriate CAPM design and implementation methodology to address the issues and concerns identified as significant. A clear distinction is made in the thesis between the work carried out by the author as part of the ACME team and the author's own work. The results from the joint effort of the research team are explained together with the author's unique contribution. The concepts of operational performance envelopes and contextually bound computing and information systems infrastructures provide the theoretical foundation to the author's approach. These concepts are utilised within an approach developed by the author which offers a wider ranging approach than is currently available. Existing approaches focus on the development of single point solutions and aim to address particular and current problems only. Such solutions are inappropriate where requirements are subject to rapid and frequent change, as in the manufacturing sector. In contrast the author's approach focuses on the development of a computing and information systems capability with the necessary flexibility to accommodate changing requirements and priorities. In this way a more resilient solution is obtainable.

The practical motivation for the work described in this thesis arose from the development of a new Jaguar car engine. Development tests on prototype engines led to multiple failure time data which are modelled as a non-homogeneous Poisson process in its log-linear form. Initial analysis of the data using failure time plots showed considerable differences between prototype engines and suggested the use of models incorporating random effects for the engine effects. These models were fitted using the method of maximum likelihood. Two random effects have been considered: a proportional effect and a time dependent effect. In each case a simulation study showed the method of maximum likelihood to produce good estimates of the parameters and standard errors. There is also shown to be a bias in the estimate of the random effect, especially in smaller samples. The likelihood ratio test has been shown to be valid in assessing the statistical significance of the random effect, and a simulation exercise has demonstrated this in practical terms. Applying this test to the models fitted to the Jaguar data gives the proportional random effect to be significant while the time dependent random effect is not found to be significantly different from zero. This test has also been demonstrated to be of use in distinguishing between the two models and again the proportional random effect model is found to be more suitable for the Jaguar data. Residual analysis is performed to aid model validation Covariates are included, in various forms, in the proportional random effect model and the inclusion of these in the time dependent model is briefly discussed. The use of these models is demonstrated for the Jaguar data by including the type of test an engine performed as a covariate. The covariate models have also been used to compare engine phases. A framework for extending the models for interval censored data is developed. Finally this thesis discusses possible extensions of the work summarised in the previous paragraphs. This includes work on alternative models, Bayesian methods and experimental design.

Changes in the competitive environment have strongly influenced manufacturing companies to adopt and develop best practice. Best practice is usually imported into companies using the services of consultancy organisations. The use of consultancy services does not guarantee success however, and inadequate results have been obtained by practitioners who have engaged in client-consultant relationships. The inadequacy of these results may be explained by the installation of pre-defined solutions by consultants as opposed to the adaptation and implementation of solutions to meet the specific requirements of practitioners. Tills may in part be explained by a lack of understanding of 'best practice'. Tills work presented in this thesis investigated the feasibility of computer based mechanisms for intervention in small and medium sized enterprises (SMEs) for the delivery of best practice. The research was undertaken using a prototyping approach. Three prototype computer based tools (CBTs) were developed by the author and tested by practitioners. The prototypes were designed based on a set of objectives and a framework of features which was developed. These frameworks were constructed from a synthesis of the research findings which included a study of best practice, the identification of characteristics of types of intervention, the identification of SME characteristics, and inhibitors of change in SMEs. The research has indicated that an approach using computer based tools is appropriate for intervention in SMEs and for adapting best practice to meet specific requirements. A structured project management approach is required with identifiable goals and benefits. An exploratory learning environment should be used to deliver complex best practice concepts and to support the goal oriented approach. Tools and techniques provided by the CBT enable the achievement of methodological tasks and facilitate experimentation and learning. The approach should not prescribe solutions, but should provide information through computer generated analyses to support decision making. The research suggests that the proposed approach may support a workbook based methodology, or may encapsulate a process methodology. The originality of this work is in the provision of a definition of best practice, an explanation of the deficiencies of existing mechanisms for the transfer of best practice to SMEs, and the specification of the features required by a new computer-based approach. Tills provides new knowledge for the field of production and operations management.

This work investigated the use of generic models in the early stages of the design and implementation of computer aided production management (CAPM) systems. A set of issues that affect the success of such CAPM systems was identified, using information obtained from literature and observations made by the author during an in-depth case study of the design and implementation of a CAPM system. The set of issues included the failure of many manufacturing companies to take a systemic perspective of CAPM and the importance of developing a thorough understanding of existing systems and how these systems are currently integrated. Requirements were proposed for an improved approach to the design and implementation of CAPM systems. Taking the requirements into consideration the concepts underlying the use of and types of models were explored. In particular, the use of generic models and how generic models could help manufacturing companies were considered. The work also investigated the use of soft systems thinking and the concept of a business process to encourage a systemic perspective to be taken. A genetic process model was proposed as a means of meeting the requirements of an improved approach. A generic model of an "order fulfilment" process in a manufacturing company was developed and a way of using it which embodies soft systems principles was proposed. The model and its use was validated using five key needs of practitioners. The validation involved a review of the model by practitioners and the use of the model in a local company as part of a project to design and implement a CAPM system. The originality of this work lies in the development by the author of a generic model which can be used as part of an improved model based approach to the design and implementation of CAPM systems. This should provide clear advantages over existing approaches.

No description available.

This work documents an investigation of the degradation of a variety of different tools whilst conducting milling operations on a computer numerical controlled (CNC) milling machine. The potential of a range of sensors to detect tool degradation has been investigated and the outputs have been incorporated into a monitoring system. Progressive degradation under nominal rough and finish face milling and rough groove milling has been investigated using a two point grooving tool and four and eight point face milling tools on En8, En24 and En24T workpiece materials. Rapid degradation of the cutting tool has also been observed under rough milling conditions using four and eight point face milling tools, whilst machining n8 and En24T materials in a variety of simulated and actual tool breakage situations. A limited investigation of the effect of the individual wear geometries associated with both progressive and instantaneous tool degradation has been conducted by simulating these geometries and carrying out rough miffing tests using a four point face milling tool on a workpiece of En8 material. Similarly, a limited investigation of the effect of machining on different machines has also been undertaken. A number of different sensing technologies have been used, including conventional sensors such as spindle current and cutting force but also novel sensing techniques such as Acoustic Emission. These have been combined using artificial intelligence techniques to provide automatic recognition of the tool wear state. Similarly, the feasibility of breakage detection/prediction has also been demonstrated.

No description available.